Information Risk and Fair Values: An Examination of Equity Betas


  • We thank the following individuals for useful discussions and comments: Mary Barth, Anne Beatty, John Core, Merle Erickson (editor), David Harris, Robert Merton, Jim Ohlson, Devin Shanthikumar, Kumar Shivakumar, Irem Tuna, Florin Vasvari, Sean Wang, and an anonymous reviewer. We also thank seminar participants from Boston University, the Financial Accounting Standards Research Initiative, the Financial Economics and Accounting 2009 Conference, the 2010 IMO Conference at Harvard Business School, the Journal of Accounting, Auditing and Finance 2009 Conference at New York University, the London Business School, New York University, Syracuse University, University of Connecticut, and the Research Accounting Conference at Yale University.


Using a sample of U.S. financial institutions, we exploit recent mandatory disclosures of financial instruments designated as fair value level 1, 2, and 3 to test whether greater information risk in financial instrument fair values leads to higher cost of capital. We derive an empirical model allowing asset-specific estimates of implied betas, and find evidence that firms with greater exposure to level 3 financial assets exhibit higher betas relative to those designated as level 1 or level 2. We further find that this difference in implied betas across fair value designations is more pronounced for firms with ex ante lower-quality information environments: firms with lower analyst following, lower market capitalization, higher analyst forecast errors, or higher analyst forecast dispersion. Overall, the results are consistent with a higher cost of capital for more opaque financial assets, but also suggest that differences in firms' information environments can mitigate information risk across the fair value designations.